ABSTRACT
BACKGROUND: Optic neuritis (ON) is a complex clinical syndrome that has diverse etiologies and treatments based on its subtypes. Notably, ON associated with multiple sclerosis (MS ON) has a good prognosis for recovery irrespective of treatment, whereas ON associated with other conditions including neuromyelitis optica spectrum disorders or myelin oligodendrocyte glycoprotein antibody-associated disease is often associated with less favorable outcomes. Delay in treatment of these non-MS ON subtypes can lead to irreversible vision loss. It is important to distinguish MS ON from other ON subtypes early, to guide appropriate management. Yet, identifying ON and differentiating subtypes can be challenging as MRI and serological antibody test results are not always readily available in the acute setting. The purpose of this study is to develop a deep learning artificial intelligence (AI) algorithm to predict subtype based on fundus photographs, to aid the diagnostic evaluation of patients with suspected ON. METHODS: This was a retrospective study of patients with ON seen at our institution between 2007 and 2022. Fundus photographs (1,599) were retrospectively collected from a total of 321 patients classified into 2 groups: MS ON (262 patients; 1,114 photographs) and non-MS ON (59 patients; 485 photographs). The dataset was divided into training and holdout test sets with an 80%/20% ratio, using stratified sampling to ensure equal representation of MS ON and non-MS ON patients in both sets. Model hyperparameters were tuned using 5-fold cross-validation on the training dataset. The overall performance and generalizability of the model was subsequently evaluated on the holdout test set. RESULTS: The receiver operating characteristic (ROC) curve for the developed model, evaluated on the holdout test dataset, yielded an area under the ROC curve of 0.83 (95% confidence interval [CI], 0.72-0.92). The model attained an accuracy of 76.2% (95% CI, 68.4-83.1), a sensitivity of 74.2% (95% CI, 55.9-87.4) and a specificity of 76.9% (95% CI, 67.6-85.0) in classifying images as non-MS-related ON. CONCLUSION: This study provides preliminary evidence supporting a role for AI in differentiating non-MS ON subtypes from MS ON. Future work will aim to increase the size of the dataset and explore the role of combining clinical and paraclinical measures to refine deep learning models over time.
ABSTRACT
The role of ICOS in antitumor T cell responses and overall tumor progression has been controversial. In this study, we compared tumor progression in mice lacking ICOS selectively in regulatory T (Treg) cells or in all T cells. Using an experimental melanoma lung metastasis model, we found that Treg cell-specific ICOS knockout reduces the overall tumor burden compared with Cre control mice, with increased CD4+-to-Treg cell and CD8+-to-Treg cell ratios in the tumor. In contrast, there was no difference in the tumor burden in mice lacking ICOS in all of the T cell compartments. This suggests a dual role of ICOS costimulation in promoting protumor and antitumor T cell responses. Consistent with reduced tumor burden, we found that Treg cell-specific deletion of ICOS leads to an increase of CD8+ CTLs that express high levels of granzyme B and perforin. Moreover, single-cell transcriptome analysis revealed an increase of Ly108+Eomeshi CD8+ T cells at the cost of the Ly108+T-bethi subset in Treg cell-specific knockout mice. These results suggest that ICOS-expressing Treg cells suppress the CTL maturation process at the level of Eomes upregulation, a critical step known to drive perforin expression and cytotoxicity. Collectively, our data imply that cancer immunotherapies using ICOS agonist Abs may work better in Treg cell-low tumors or when they are combined with regimens that deplete tumor-infiltrating Treg cells.